CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 51, 2016 

A publication of 

 
The Italian Association 

of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors: Tichun Wang, Hongyang Zhang, Lei Tian 
Copyright © 2016, AIDIC Servizi S.r.l., 

ISBN 978-88-95608-43-3; ISSN 2283-9216 

Optimizing the Schedule of Dispatching Construction 
Machines through Artificial Intelligence 

Yan Xing*, Zhibin Song, Xilu Deng 
*College of Urban Construction , Hebei Normal University of Science and Technology, Qin Huangdao ,Hebei, China,066400  
xingyanqq@163.com 

Most construction projects involve use of construction machines, construction plant manager has to consider 
both timeliness and flexibility to develop an efficient schedule of dispatching construction machines, which can 
balance the operations at different construction sites. In this paper, we constructed an optimization model with 
multi-target as construction quality, time limit, efficiency and cost. The model consists of a database, a rule-
based system and multi-target decision making modules. The database includes detailed data about 
construction machine types and their properties. The rule-based system module provides rules, which are 
utilized by inference engine for determining the most proper construction machine type. Ultimately, a final 
decision is made for the most proper construction machine among the alternatives of the same type using the 
information axiom of axiomatic design principles. In order to verify the validity of the model, evaluation of 
alternatives is made for the cases of both complete and incomplete information. 

1. Introduction 

Construction machine selection is an important activity in the design of an effective construction operation 
system design. Dispatching appropriate construction machine can decrease constructing times, improve 
construction machine utilization and increase quantities. Construction machine operating can account for 30–
55% of the total cost, and efficient construction machine operating can be primarily responsible for reducing a 
construction site’s operating cost by 15–30%. Therefore, determination of proper construction machine 
dispatching strategy is very important for reduced costs and increased profits. As a wide variety of 
construction machine is available today, each having distinct characteristics and cost that distinguish from 
others, determination of the proper construction machine for a designed construction operation is a very 
complicated decision.  
There are various approaches focused on the solution of this complicated problem. Intelligent computer 
systems have been developed such as expert systems and decision support systems for the dispatching of 
construction machine. 
In artificial intelligence, an expert system is a computer system that emulates the decision-making ability of a 
human expert. Expert systems are designed to solve complex problems by reasoning about knowledge, 
represented primarily as if–then rules rather than through conventional procedural code. One of the most 
successful applications of experts systems is SEMH: selection of equipment for material handling. SEMH 
searches its knowledge base to recommend the degree of mechanization, and the type of material handling 
equipment to be used, based on some characteristics (Fonseca et al., 2004). Malmborg (Malmborg et al., 
1987) has developed a prototype expert system considering 17 equipment attributes and 47 devices for 
industrial truck type selection. 
A decision support system (DSS) is a computer-based information system that supports business or 
organizational decision-making activities. DSSs serve the management, operations, and planning levels of an 
organization and help people make decisions about problems that may be rapidly changing and not easily 
specified in advance. Decision support systems can be either fully computerized, human-powered or a 
combination of both. DSSs include knowledge-based systems. A properly designed DSS is an interactive 
software-based system intended to help decision makers compile useful information from a combination of 

                               
 
 

 

 
   

                                                  
DOI: 10.3303/CET1651083

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Xing Y., Song Z.B., Deng X.L., 2016, Optimizing the schedule of dispatching construction machines through artificial 
intelligence, Chemical Engineering Transactions, 51, 493-498  DOI:10.3303/CET1651083   

493



raw data, documents, and personal knowledge, or business models to identify and solve problems and make 
decisions. 
This work is organized as follows. Section 2 describes related work. In Section 3 we explain factors of 
impacting the schedule of dispatching construction machines. In Section 4 structure of optimal scheduling 
model is introduced, and the probability equation of success by integrating the system probability density 
function over the entire design range is given. In Section 5 evaluation method of alternatives is made for the 
cases of incomplete information is proposed. In Section 6 we conclude this paper. 

2. Related work 

Construction is one of the most common and essential operations in civil engineering projects, its procedures 
deal mainly with earth excavation, transportation and dumping, all of which require construction machine 
operations. Considerable work has been done developing models or approaches to facilitate the planning of 
construction machine operations. 
Marzouk and Moselhi (Marzouk et al., 2003) have provided a simulation model to optimize the use of available 
resources. Moselhi and Alshibani (Moselhi et al., 2009) have proposed an optimization of earthmoving 
operations, which focused on selecting suitable machine equipment. Jayawardane and Harris (Jayawardane 
et al., 1990) proposed a method which was used to minimize cost while completing the projects within the 
targeted time frame. Son and Mattila (Son et al., 2005) propose a model which was used to balance the 
utilization of resources throughout the project duration. 
Existing works for dispatching construction machines have several limitations. They always have been done 
from the general contractors’ point of view. In other words, the previously developed models are designed to 

dispatch construction machines for only a single project.  
Most of these models or approaches do not consider all of the technical, strategic and economic criteria 
simultaneously in dispatching the most appropriate construction machine. They also cannot evaluate the 
alternatives for the cases of both complete and incomplete information. 
Approaches that include more than one measure of performance in the evaluation process are termed multi-
attribute or multi-criteria decision methods. The advantage of these methods is that they can account for both 
financial and non-financial impacts. 
The main goals and contributions of this paper are the use of computer simulation methods and axiomatic 
design approach, with the objective of minimizing the time for constructing an optimal dispatch model to deal 
with construction machines schedule problems. 

3. Factors of impacting the schedule of dispatching construction machines 

There are many factors that impact the schedule of dispatching construction machines, the main factors are 
as follows: 

3.1 Travel time between the construction site and the landfill site 

The travel time between the construction site and the landfill site is not easy to predict exactly, as it is 
determined by the distance between the locations, the speed of the transportation machines, and traffic 
conditions. To avoid delays, the dispatching manager usually assigns a higher priority to job sites that have a 
large volume of earth to be removed. However, such an approach increases the chance of interrupting the 
work process at a job site that has a faster but lower-volume earthmoving operation. Therefore, the travel time 
between the job site and landfill site is important when deciding the schedule of the construction machines. In 
practice, the average travel time can be estimated from historic data. 

3.2 Working time at the job site 
The time spent on working at the job site varies with the type of construction activity, which could affect the 
frequency with which machines are dispatched to the same job site. For example, the faster the constructing 
operation at a job site, the shorter is the interval between dispatches of machines to that job site. If the 
construction machines cannot arrive at the job site in time, it is possible that the whole job site will be idle, 
which will impact the work process at the job site tremendously. The dispatching manager may assign as 
many construction machines as possible to the busiest job site; however, such an approach might cause time 
wastage if the construction machines need to line up and wait for working. 

3.3 Number of construction machines needed 
The number of construction machines needed depends on the amount of earth-work scheduled to be removed, 
the loading capacity of the construction machines and the road’s regulated weight limit. For example, one 
construction site schedules 300 m3 of earth-work to be removed. Although a truck can be loaded up to 20 m3, 
the regulation for the road it must travel may only permit the truck to be loaded with 15 m 3. As a result, the 

494



number of trips that trucks make to this job site increases from 15 to 20, and the time needed to finish the 
earthmoving process is extended. 

4. Structure of optimal scheduling model 

The optimal scheduling model is developed in order to simplify the selection process of the most appropriate 
construction machines, consists of the following modules as shown in Figure 1. 

 

Figure 1: Flow-process diagram of the optimal scheduling model 

Construction machines are classified into main groups of trucks, excavators, cement mixers, cranes, forklift 
and boring machines. This module includes examples of 46 move construction machine types with their 
performance criteria. 
The model includes 137 rules in the knowledge base, assuming a 100% confidence factor. These rules are 
acquired from construction machines dispatching managers and the literature about construction machines. 
Arity PROLOG language is used to define these rules so that they will not be badly influenced by the 
variations in the methods or the changes in the knowledge base. Some examples of knowledge base rules, 
which are in accordance with IF–THEN structure, are presented below. 
Rule 37: 
 
IF Function type_move and 

Move area and path_fixed and 

Floor space_available 

THEN Most proper move type_conveying. 

 

495



Rule 78: 
 
IF Move type_conveying and 

Individual type_packaged and 

Move direction_horizontal and 

Operation control_uncontrollable and 

Bottom surface_not flat 

THEN Most proper conveyor type_Wheel conveyor. 

 
Rule 79: 
 
IF Move type_conveying and 

Individual type_packaged and 

Move direction_horizontal and 

Operation control_controllable and 

Bottom surface_not flat and 

Material weight_X!100 kg 

THEN Most proper conveyor type_Belt conveyor. 

 
The most important concept in axiomatic design is the existence of the design axioms. The first design axiom 
is known as the independence axiom and the second axiom is known as the information axiom. 
The independence axiom states that the independence of functional requirements (FRs) must always be 
maintained, where FRs are defined as the minimum set of independent requirements that characterizes the 
design goals (Suh, 1995). In the real world, engineers tend to tackle a complex problem by decomposing it 
into sub-problems and attempting to maintain independent solutions for these smaller problems. This calls for 
an effective method that provides guidelines for the decomposition of complex problems and independent 
mappings between problems and solutions. 
The information axiom states that among those designs that satisfy the independence axiom, the design that 
has the smallest information content is the best design (Suh et al., 1998). 
Information is defined in terms of the information content, Ii, that is related in its simplest form to the probability 
of satisfying the given FRs. Ii determines that the design with the highest probability of success is the best 
design. Information content Ii for a given FRi is defined as Eq(1). 

2

1
log ( )

i

i

I
p

  (1) 

where pi is the probability of achieving the functional requirement FRi and log is either the logarithm in base 2. 
The definition of information follows the definition of Shannon, although there are operational differences. 
Because there are n FRs, the total information content is the sum of all these probabilities. When all 
probabilities are one, the information content is zero, and conversely, the information required is infinite when 
one or more probabilities are equal to zero.  
In any design situation, the probability of success is given by what designer wishes to achieve in terms of 
tolerance (design range) and what the system is capable of delivering (system range). 
As shown in Figure 2, the overlap between the designer-specified ‘design range’ and the system capability 
range ‘system range’ is the region where the acceptable solution exists. Therefore, in the case of uniform 
probability distribution function pi may be written as Eq(2). 

 
( )

 
i

Common range
p

System range
  (2) 

Therefore, the information content is equal to 

2

 
log ( )

 
i

System range
I

Common range
  (3) 

The probability of achieving FRi in the design range may be expressed, if FRi is a continuous random variable, 
as 

1
( )

u
dr

i s i i
dr

p p FR dFR   (4) 

496



( )
s i

p FR  is the system probability density function for FRi. 
Eq. (4) gives the probability of success by integrating the system probability density function over the entire 
design range. 

 

Figure 2: Design range, system range, common range and probability density function of a FR 

In Figure 3, the area of the common range (Acr) is equal to the probability of success pi. Therefore, the 
information content is equal to 

2

1
log ( )

cr

I
A

  (5) 

 

Figure 3: Area within common range 

5. Fuzzy information axiom for incomplete information 

The crisp information axiom approach can be used for the solution of decision-making problems under 
certainty. This approach cannot be used with incomplete information, since the expression of decision 
variables by crisp numbers would be ill defined. The definition and formulation of the developed fuzzy 
approach are given in the following. 
The data relevant to the criteria under incomplete information can be expressed as fuzzy data. The fuzzy data 
can be linguistic terms, fuzzy sets, or fuzzy numbers. Linguistic terms are first transformed into fuzzy numbers. 
Then these fuzzy sets are assigned crisp scores. The following numerical approximation system are proposed 
to systematically convert linguistic terms to their corresponding fuzzy numbers. 
In the fuzzy case, we have incomplete information about the system and design ranges. The system and 
design range for a certain criterion will be expressed by using ‘over a number’, ‘around a number’ or ‘between 
two numbers’. Triangular or trapezoidal fuzzy numbers can represent these kinds of expressions. We now 
have a membership function of triangular or trapezoidal fuzzy number (TFN), whereas we have a probability 
density function in the crisp case. 
The common area, which is the intersection area of triangular or trapezoidal fuzzy numbers, between design 
range and system range is shown in Figure 4. 

497



 

Figure 4: The common area of system and design ranges 

6. Conclusions 

A schedule that satisfies the requests from different construction operations is needed to quickly generate for 
dispatch planning manager. However, completing such an operation is not an easy task and, in practice, 
depends mainly on the experience of the manager. In this paper a decision support model, which considers 
both technical and economic criteria in dispatching construction machines process, is presented. The rule-
based system module provides rules, which are utilized for determining the most proper construction machine 
type. Ultimately, a final decision is made for the most proper construction machine among the alternatives of 
the same type using the information axiom of axiomatic design principles. 

Reference  

Fonseca D., Uppal G., Greene J., 2004, A knowledge-based system for conveyor equipment selection, Expert 
Systems with Applications, 26, 615- 623, DOI: doi:10.1016/j.eswa.2003.12.011 

Jayawardane K., Harris C., 1990, Further development of integer programming in earthwork optimization, 
Journal of Construction Engineering and Management, 1, 18-34, DOI: 10.1061/(ASCE)0733-
9364(1990)116:1(18) 

Malmborg J., Agee H., Choudhary V., 1987, A prototype expert system for industrial truck type selection, 
Industrial Engineering, 3, 58-65  

Marzouk M., Moselhi O., 2003, Object-oriented simulation model for earthmoving operations, Journal of 
Construction Engineering and Management, 2, 173-181, DOI: 10.1109/WSC.2000.899188M 

Moselhi O., Alshibani A., 2009, Optimization of earthmoving operations in heavy civil engineering projects, 
Journal of Construction Engineering and Management, 10, 948 -954, DOI: 10.1061/(ASCE)0733-
9364(2009)135:10(948) 

Son, J., Mattila K., Myers D., 2005, Determination of haul distance and direction in mass excavation, Journal 
of Construction Engineering & Management, 3, 302-309, DOI: 10.1061/(ASCE)0733-
9364(2006)132:8(897) 

Suh, P., 1995, Designing-in of quality through axiomatic design, IEEE Transactions on Reliability, 2, 256-264, 
10.1109/24.387380 

Suh, P., Cochran, S., Paulo, L., 1998, Manufacturing system design, Annals of the CIRP, 2, 627-639, DOI: 
10.1016/S0007-8506(07)63245-4 

498